Tilting Together: An Information-Theoretic Characterization of Behavioral Roles in Rhythmic Dyadic Interaction

Area of Science:

• Neuroscience
• Human-Computer Interaction
• Social Robotics

Background:

• Joint physical activities necessitate spatiotemporal coordination of individual motor behaviors.
• Interpersonal coordination can emerge from observing partners' and object movements, bypassing verbal communication.
• Understanding social coupling is key to analyzing collaborative task performance.

Purpose of the Study:

• To investigate the translation of social coupling into objective and subjective performance indicators in collaborative tasks.
• To analyze dyadic interrelationships using transfer entropy to identify leader-follower roles.
• To explore causal dependencies in social interaction for insights into sensorimotor contingencies.

Main Methods:

• Experimental paradigm involving joint control of a simulated object (ball) on a tablet.
• Subjects synchronized movements with a 90° phase difference to maintain specific trajectories.
• Information-theoretic measure of transfer entropy used to analyze dyadic interrelationships.

Main Results:

• Demonstrated emergence of spatiotemporal coordination in collaborative physical activity.
• Identified leader-follower roles through analysis of transfer entropy.
• Revealed specific trends in human behavior and social sensorimotor contingencies.

Conclusions:

• Causal dependencies in social interaction tasks offer insights into human behavior.
• Objective and subjective performance indicators reflect social coupling.
• Movement observation is sufficient for emergent interpersonal coordination.